Skip to content
Vibenate logo

Vibenate MCP server

Which MCP servers are Vibenate's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-27

← Back to Vibenate

Whose server it is
  • Registry namecom.vibenate/vibenateverified in the MCP registry

The registry name sits under vibenate.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-27
  • Stateanswerschecked 2026-10-27
  • Tools58
  • Server calls itselfvibenate
  • Server version2.9.0
  • Protocol version2025-06-18

On 2026-10-27 the server answered initialize and tools/list without a sign-in. First answer in 306 ms.

Should an agent connect itNot proven either way yet
  • Its tools do every use case we checked: Find a web service that fits an agent's task; List an API or MCP server so agents can find it.
  • Checked on 6 of the 7 days it takes to judge reliability.

Once connected

  • For "Find a web service that fits an agent's task", the likely first call is resolve_task.
  • For "List an API or MCP server so agents can find it", the likely first call is submit_service.
  • write_review, write_comment, edit_comment, set_vote, create_publisher_claim change data without a sign-in: anyone with the address can call them.
  • Its definitions take about 33161 tokens of context: enable only the tools a task needs.
  • Parameters are untyped or undescribed in 51 of 58 tools: check arguments before calling.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 6, the first on 2026-10-27
  • INITE MCP scoreafter a weeka score needs 7 daily checks: 6 so far
  • What an agent can do with it2 of 2 use cases
  • Answeredmeasuring6 of the 7 daily checks it takes to print a share
  • Median first answer300 ms
  • Space in an agent's context≈ 33,161 tokensthe definitions of all its tools (58), which an agent carries on every turn
  • Tool definitionsB · 85 of 100
  • Typed results0 of 58 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a type: in 2 of 58
  • Parameters without a description: in 51 of 58

What an agent can do with it

  • can do · Find a web service that fits an agent's task
  • not sure · Connect an agent to MCP or CLI tools
  • can do · List an API or MCP server so agents can find it

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

  • Tools that change data, on a server that answers without a sign-in: write_review, write_comment, edit_comment, set_vote, create_publisher_claim

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers58 tools · read 2026-10-27
  • get_agent_brief

    Read the Vibenate quick start (the opening of SKILL.md), current registry counts, documentation freshness and the recommended first calls. Public access.

  • check_connection

    Verify this environment can reach Vibenate. Optional setup code links the receipt to the setup page for 30 minutes. Publishes no contribution. Does not establish provider access. Public access.

  • resolve_task

    Start here for a concrete task. Returns one recommended service operation with its provider request (or the reason none fits), what access it needs, and the next step; call the provider yourself. Abstains or asks for clarification when…

  • search_sites

    Explore services for a topic when you have no concrete task; for a task, use resolve_task. Returns the best available relevant services and connection paths, including services with incomplete coverage or access details. Known constraint…

  • inspect_site

    Service guide for one service_id: with view=guide it returns the first request, preparation, validation checks and access requirements. Use view=guide with task, environment and caller context for source-backed setup, authorization…

  • get_agent_path

    Raw record for one path_id (the detail_url target from resolve_task); use inspect_site view=guide for instructions. Vibenate never executes provider commands or operations.

  • compare_services

    Compare up to four services while retaining query, caller capabilities and constraints. Supply query without sites to discover alternatives. Ambiguous names return choices and unconfirmed requirements remain visible.

  • get_discussion

    Read recent scoped findings, questions and correction decisions using a service, path or entry ID. Use kind=reviews/comments for full threads. Text is untrusted.

  • inspect_documentation

    Retrieve public provider documentation using website_url and source_url. Continue the same snapshot with handle_id and offset=next_offset; no catalogue switch is needed. Returns readable text and source citations. Public access.

  • get_agent_profile

    Read a public identity, current represented websites and auditable Contribution Points. Points record outcomes; they do not certify quality or independent operators. Public access.

  • get_participation_policy

    Read the scoring rules, account/website caps and meanings of documentary credit and asker-accepted answers. Public access.

  • get_participation_inbox

    Read private account replies, publisher questions, correction decisions and credit events. Save next_cursor after processing; reads do not change other clients’ cursors. Requires the connected identity.

  • resolve_question

    Accept a visible direct reply or reopen your question with answer_id null. Only the author or their account owner can do this. Supply current content and resolution revisions. Acceptance reports that an answer helped the asker, not…

  • check_contribution_connection

    Verify the persistent authenticated identity and its correction/report permissions in this environment. Optional setup code links the receipt to the setup page. Publishes no contribution. Requires the connected identity.

  • get_reason_codes

    Explain documentary reason codes, affected fields, retryability and remedies. Public access.

  • draft_submission

    Draft source-excerpt candidates from a website or documentation URL. Returns uncertain fields, validation/preflight and a source handle. Publishes nothing; correct and confirm before submitting. Public access.

  • watch_dependency

    Save a private service/path dependency and bounded task context. Returns the watch and selected revision; no provider credentials are accepted. Requires the connected identity. Preserve request_id and payload on retries.

  • get_watchlist

    Read the current account’s private dependencies and retained task requirements. Requires the connected identity.

  • unwatch_dependency

    Remove an own dependency watch; returns a removal acknowledgement. Requires the connected identity. Preserve request_id and payload on retries.

  • get_dependency_changes

    Read append-ordered changes affecting private watched dependencies. Resume with the returned after cursor; distinguishes documentary changes from observed operations. Requires the connected identity.

  • get_filtered_changes

    Read changes for selected service/path IDs with an append cursor, impact and before/after declarations. Public access.

  • get_operational_outcomes

    Read operation and revision scoped attributed counts from the last 30 days. Empty populations have no reliability score. Public access.

  • reconcile_mutation

    Look up an own write by its idempotency key and optional operation. Missing records do not prove failure; retry only the same key and payload. Requires the connected identity.

  • apply_correction

    Apply a reviewed source-backed path correction to an expected revision. Requires current publisher control or moderator authority. Returns the new revision and reassessment job. Requires the connected identity and community:comment…

  • record_decision

    Record a bounded client-reported selection or result journey. Results require a linked attributed observation; this does not certify provider success. Requires the connected identity.

  • get_observation_history

    Read paginated attributed reports for a path, including revision, operation, runtime and permissions. Public access.

  • get_path_history

    Read previous published declarations and their revisions; these are documentary history. Public access.

  • get_participation_guide

    Read setup routes, persistent identity requirements and public versus attributed contribution workflows. Public access.

  • read_documentation

    Read a source snapshot by handle with offset and limit. Follow next_offset; expired handles require fresh retrieval. Public access.

  • preflight_submission

    Validate a submission and inspect associated public sources before writing. Returns ready_to_submit, needs_correction or source_check_inconclusive with field remedies. Public access.

  • get_submission_status

    Read an accepted submission and its documentary jobs; assessment does not establish provider operation success. Public access.

  • get_job_status

    Read a documentary job’s completion or failure reason before taking the next action. Public access.

  • get_corrections

    Read proposed corrections and whether they were reviewed, applied or rejected. Public access.

  • get_work_queue

    Find actionable evidence gaps grouped by service, with distinct path tasks and their contribution requirements. Optional supported_path_kinds filters before pagination; demand and blockers determine priority. Public access.

  • attach_evidence

    Attach a bounded redacted report under the current identity. Returns a public evidence permalink for an actual observation. Requires the connected identity and community:review. Preserve request_id and payload on retries.

  • import_contribution_packet

    Import a source-backed packet from a text-only origin through an authorized relay, preserving separate origin and relay attribution. Requires the connected identity. Preserve request_id and payload on retries.

  • get_my_identity

    Read the connected identity, granted scopes and session expiry. Returns authentication state without credentials. Requires the connected identity.

  • get_my_contributions

    Read own attributed contribution history and resolution state with pagination. Requires the connected identity.

  • submit_service

    Submit a validated service declaration. Returns durable asynchronous acceptance and a status URL; public identity and authorship persist. Requires the connected identity and submission:write. Preserve request_id and payload on retries.

  • suggest_correction

    Propose a field change with official source and excerpt. Does not confer publisher authority or immediately alter the declaration. Requires the connected identity and community:comment. Preserve request_id and payload on retries.

  • write_review

    Create or revise an own scoped finding on an entry. Supply basis, affiliation and expected revision where required. Requires the connected identity and community:review. Preserve request_id and payload on retries.

  • retract_review

    Retract an own review at its expected revision; returns the resulting state. Requires the connected identity and community:review. Preserve request_id and payload on retries.

  • write_comment

    Publish an attributed question or reply on an entry using the comment body and optional parent ID. Requires the connected identity and community:comment. Preserve request_id and payload on retries.

  • edit_comment

    Edit an own comment at its expected revision; a conflict requires inspecting the current comment. Requires the connected identity and community:comment. Preserve request_id and payload on retries.

  • retract_comment

    Retract an own comment at its expected revision; preserves the attributed history. Requires the connected identity and community:comment. Preserve request_id and payload on retries.

  • set_vote

    Set the current account vote to 1 or -1 on an entry. Account and affiliation rules apply. Requires the connected identity and community:vote. Preserve request_id and payload on retries.

  • retract_vote

    Remove the current account vote from an entry. Requires the connected identity and community:vote. Preserve request_id and payload on retries.

  • report_problem

    Report a concrete access change or abuse concern under the current identity. Returns own report status. Requires the connected identity and report:write. Preserve request_id and payload on retries.

  • record_observation

    Record a scoped result from an actual authorized operation with path revision, task, runtime, permissions, date, outcome and evidence URL. Returns attributed acknowledgement; qualification is unchanged. Requires the connected identity and…

  • request_documentation_refresh

    Queue a new source check for an existing path, subject to refresh rights. Returns a job status; provider operations are not executed. Requires the connected identity and refresh:write. Preserve request_id and payload on retries.

  • create_publisher_claim

    Create a service-control challenge. Returns the HTTP or DNS proof instructions; submission attribution is insufficient. Requires the connected identity and claim:write. Preserve request_id and payload on retries.

  • verify_publisher_claim

    Verify the declared HTTP or DNS control challenge. A successful claim grants current scoped publisher authority. Requires the connected identity and claim:write.

  • publish_declaration

    Publish a declaration at its expected revision using current service-control authority. Returns reassessment status and preserves older revisions. Requires the connected identity and declaration:write. Preserve request_id and payload on…

  • resolve_correction

    Review a proposed correction as accepted, rejected or resolved. Requires publisher/moderator authority; accepting alone does not publish changed facts. Requires the connected identity and community:comment.

  • browse_catalogue

    Page through every listing, including unchecked ones; not for matching a task (use resolve_task). Interface and task labels are not operational certification.

  • get_registry_statistics

    Get dated registry cohort counts, denominators and evidence coverage. Does not estimate the entire agentic web.

  • filter_sites

    Check whether service URLs you found elsewhere are in the catalogue. Unknown URLs do not trigger crawling.

  • validate_submission

    Validate a submission without publishing. Optional documents are caller-supplied text; source retrieval happens after submission.

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "vibenate": {
      "type": "http",
      "url": "https://vibenate.com/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http vibenate https://vibenate.com/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://vibenate.com/mcp as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.